Papers › Learning Language-guided Adaptive Hyper-modality Representation for Multimodal...

Learning Language-guided Adaptive Hyper-modality Representation for Multimodal Sentiment Analysis

9 Oct 2023arXiv:2310.05804archive 2025-07-28

Haoyu Zhang, Yu Wang, Guanghao Yin, Kejun Liu, Yuanyuan Liu, Tianshu Yu

Though Multimodal Sentiment Analysis (MSA) proves effective by utilizing rich information from multiple sources (e.g., language, video, and audio), the potential sentiment-irrelevant and conflicting information across modalities may hinder the performance from being further improved. To alleviate this, we present Adaptive Language-guided Multimodal Transformer (ALMT), which incorporates an Adaptive Hyper-modality Learning (AHL) module to learn an irrelevance/conflict-suppressing representation from visual and audio features under the guidance of language features at different scales. With the obtained hyper-modality representation, the model can obtain a complementary and joint representation through multimodal fusion for effective MSA. In practice, ALMT achieves state-of-the-art performance on several popular datasets (e.g., MOSI, MOSEI and CH-SIMS) and an abundance of ablation demonstrates the validity and necessity of our irrelevance/conflict suppression mechanism.

Paper PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Haoyu-ha/ALMT officialpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Multimodal Sentiment Analysis Sentiment Analysis

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Multimodal Sentiment Analysis	CH-SIMS	ALMT	Acc-2	81.19	#2 of 2	Archive leaderboard	report
Multimodal Sentiment Analysis	CH-SIMS	ALMT	Acc-3	68.93	#2 of 2	Archive leaderboard	report
Multimodal Sentiment Analysis	CH-SIMS	ALMT	Acc-5	45.73	#2 of 2	Archive leaderboard	report
Multimodal Sentiment Analysis	CH-SIMS	ALMT	CORR	0.619	#2 of 2	Archive leaderboard	report
Multimodal Sentiment Analysis	CH-SIMS	ALMT	F1	81.57	#2 of 2	Archive leaderboard	report
Multimodal Sentiment Analysis	CH-SIMS	ALMT	MAE	0.404	#2 of 2	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSEI	ALMT	Acc-5	55.96	#15 of 15	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSEI	ALMT	Acc-7	54.28	#15 of 15	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSEI	ALMT	Corr	0.779	#15 of 15	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSEI	ALMT	MAE	0.526	#15 of 15	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSI	ALMT	Acc-5	56.41	#10 of 12	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSI	ALMT	Acc-7	49.42	#10 of 12	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSI	ALMT	Corr	0.805	#10 of 12	Archive leaderboard	report
Multimodal Sentiment Analysis	CMU-MOSI	ALMT	MAE	0.683	#10 of 12	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position Encodings Adam Attention BPE Dense Connections Dropout Label Smoothing Layer Normalization Linear Layer Multi-Head Attention Position-Wise Feed-Forward Layer Residual Connection Softmax Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections